ABSTRACT
Project portfolio managers rely on data-driven approach to collect and analyze data based on a combination of metrics and methods for making strategic choices, to determine how the business will look like ve years out, and particularly to balance resource allocation. e metrics used to select projects, range from quantitative (e.g., return on investment (ROI) for nancial models and nancial indices; probabilistic nancial models; option pricing theory; etc.) to qualitative (e.g., alignment with company strategy; strategic approaches; scoring models and checklists; analytical hierarchy approaches; etc.). Despite of the use of a variety of methods and metrics, projects selection and project prioritization are two of the most cited project activities that are still presenting the weakest facets respect to all the other processes involved with project portfolio management as reported in (Cooper, Edgett, & Kleinschmidt, 1998). However according to recent surveys, companies still have diculties in the selection criteria (Partners, 2013). Consequentially the poor performance in selection, prioritization, and thus in the decision-making process is reected into the misalignment and in the general poor performance of the portfolio management. Is it possible that hidden or undiscovered predictive indicators within project portfolio data, with their distributions that have been overlooked within the standard methods, could address project issues such as selection and prioritization from misalignment?
